Staff Machine learning software engineer with ~7 years experience specializing in on-device inference optimization for Qualcomm's GPU enabled hardware.I am currently the technical lead for Qualcomm's Neural Processing GPU component. I directly lead a team of 6 engineers and own a project that 15 engineers contribute to. Our team develops and optimizes OpenCL kernels running on (mostly) arm Android devices to accelerate inference times while maintaining accuracy. We triage, implement, and integrate OpenCL/GPU/general features such as optimized OpenCL memory types, node fusions, GMEM enablement, memory reduction techniques for LLMs, tensor sharing, dynamic tensors, etc. My current role includes coordinating with Product Management, Customer Engineers, and stake holders to understand requests, communicate updates and statuses, and plan road maps. I am in charge of team ticket/JIRA maintenance, release/documentation health, allocating and assigning tasks/assessing bandwidth, holding regular scrums and 1-on-1's, etc. I triage most all large projects, create documentation, tasks, timelines, and assess project needs/requirements while also taking on a handful of development work as well.My role also includes organization management in which I assist with the career development of 5 eng/sr-engs (not on my technical team). This includes: regular management syncs, goal planning, annual review, promotion assessment, etc.